PI: Madabhushi, Anant Early detection of prostatic adenocarcinoma (CAP) offers the best hope of curing it. Currently, the only way to definitively diagnose CAP is through histological analysis of excised tissue, obtained via blind sextant trans-rectal ultrasound (TRUS)-directed biopsies. However, the accuracy of TRUS biopsies is only 20-25% for elevated PSA levels. Recently, high-resolution magnetic resonance (MR) imaging (MRI) has been shown to be superior to ultrasound in visualizing prostatic structures. Computer-aided diagnosis (CAD) refers to the use of computer programs in assisting radiologists detect abnormalities from radiological images. The aim of this project is to develop a CAD system to detect prostate cancer from multi-protocol 3 Tesla in vivo MRI and Magnetic Resonance Spectroscopic (MRS) Imaging (MRSI). The initial higher costs associated with MRI will be offset by a reduction in unnecessary biopsies and an increase in the number of cancers detected via biopsy. The long term goals of this work are to motivate use of CAD with high resolution MRI/MRSI:- 1. For early detection of CAP in moderate to high risk patients and thereby reduce the number of unnecessary biopsies and radical prostatectomies. 2. Improving cancer detection rates by replacing TRUS with CAD-assisted MRI guided prostate biopsies. 3. For improved target dose conformality in prostate therapy. 4. For assessing quantitative response to anti-cancer therapy. The proposed work comprises a total of 3 specific aims and 10 tasks. A total of 40 anonymised patient data sets comprising multi-protocol 3 Tesla (T) in vivo MRI/MRSI scans before and multi-protocol 3 T ex vivo MRI/MRSI scans with accompanying whole mount histological sections after prostatectomy will be obtained (Aim 1). Four different MR protocols will be obtained for each patient study -- T1-, T2-weighted, dynamic contrast-enhanced, and diffusion imaging. The inclusion of histological data will allow for precise determination of CAP regions via H&E staining. In order to train the CAD system to automatically identify CAP on in vivo MRI accurate determination of spatial location of cancer in vivo is required. Aim 2 involves determination of spatial extent of CAP (tumor ground truth) in vivo by first registering the whole mount histological sections with the corresponding ex vivo MRI images. Once the spatial location of CAP has been thus determined on ex vivo MRI, these images are then registered with the corresponding in vivo MRI studies. Aim 3 involves building and evaluating the CAD system for detecting CAP on the high resolution in vivo MRI and MRSI studies. This project will be a collaborative effort between investigators at Rutgers University and the University of Pennsylvania (UPENN). Aim 1 (Data Generation) will be carried out at UPENN while Aims 2 (Tumor ground truth generation) and 3 (CAD model) will be done at Rutgers. 1 PI: Madabhushi, Anant Prostatic adenocarcinoma (CAP) is the second most common malignancy among men with an estimated 230,000 new cases in the USA in 2004. The current protocol to diagnose CAP in the USA is to have patients with elevated prostate specific antigen (PSA) levels undergo a trans-rectal ultrasound (TRUS)-directed biopsy. However, several studies have reported (1) that using PSA alone as a screening test results in a large number of unnecessary biopsies and (2) there is a high false negative rate associated with TRUS. The goal of this work is to integrate computer-aided diagnosis (CAD) with multi-protocol high resolution in vivo Magnetic Resonance Imaging (MRI) and Magnetic Resonance Spectroscopic Imaging (MRSI) in order to increase the sensitivity and specificity of prostate cancer detection. The proposed work has translational implications in, structures (urethra and rectal wall) by directing therapy towards specific areas determined as tumor by CAD.